The goal of this proposal is to develop network methodology to identify specific patterns of communicative interaction that predict Autism Spectrum Disorders (ASD) around a child's first birthday. At present it is difficult to diagnose ASD reliably prior to the age of two and yet parental assessment of infant communicative behavior shows promise in predicting ASD around the age of one year. Thus, there is a need to go beyond parental judgment to identify the specific communicative interactions that predict the diagnosis of ASD. In typically-developing (TD) infants, communicative development (e. g., vocal, gestural, attentive) is influenced by contingent dyadic interactions with their caregivers. These are the same communicative behaviors that are deficient in children with an ASD disorder. Thus, the sequencing of contingent dyadic interactions in children at risk for ASD could be a sensitive indicator of typical or atypical development. We propose to develop and test a new social network methodology based on the triadic census that can be used to identify the pattern of interactions (e.g., development of communicative behaviors like joint attention) that are predictive of the emergence of ASD. Triads are defined as directed interactions among up to three individuals or objects. We will use animal and a human model systems to test this new methodology. In both systems, there exist extensive data sets that can be used to develop and test the triad methodology as predictors of typical or atypical development. In the animal system, we can evaluate experimentally manipulated social groups, providing a means to describe the causal relationship between social interactions and the development of competent communicative behavior. In the human system we will use data currently being collected on infants at risk for ASD. In the human system, preliminary triadic analysis of infants at heightened biological risk for ASD revealed specific sequences of caregiver, infant and object interactions associated measures of comprehension at 14 months. PUBLIC HEALTH RELEVANCE: Relevance The purpose of this proposal is to development social network methodology to be used to detect typical and atypical communicative development in young infants. Both animal and human models will be studied. The goal is to develop a diagnostic tool with the ability to predict Autism Spectrum Disorder by the child's first birthday.